Induction of highly potent, broadly cross-reactive neutralizing antibodies is a major milestone in the pathway to development of an HIV-1 vaccine. Our laboratory has recently reported the induction of extensively cross-reactive neutralizing antibodies in rabbits, using an oligomeric form of the envelope protein (Env) ectodomain (gp140) of a particular strain of HIV-1 in a selected adjuvant for immunization. The strain of Env that was used, strain R2 is unusual in that it is from a donor with broadly cross-reactive neutralization, exhibits CD4 independent infectivity, and is neutralized broadly by antibodies directed against conformational epitopes in gp120 and gp41. The oligomeric gp140R2 binds antibodies against numerous conformational epitopes in gp120 and gp41, including mAbs against CD4-induced (CD4i) epitopes. These and other data are consistent with the hypothesis that gp140R2 expresses multiple conformational neutralization epitopes in and unmasked conformation such that they are competent for induction of cross-reactive neutralizing antibodies. In rabbits gp120R2 induced neutralizing antibodies with very limited cross-reactivity, but these antibodies developed in a sequence consistent with induction of a classical memory response and booster effect. In contrast, the highly cross-reactive neutralizing antibodies induced by gp140R2 developed much more slowly, with poor booster effect and to only modest potency. This proposal attempts to understand reasons for the limited potency of the broadly cross-reactive neutralizing response. In Aim 1 we propose to study whether insufficient stability of the gp140 trimeric quaternary structure or inadequate induction of T cell help explain the observations. In Aim 2 we pursue a series of experiments that test the hypothesis that B cells reactive with cross-reactive neutralization epitopes compete poorly in germinal center reactions with the more immunodominant epitopes presented by monomeric gp120 that induce neutralizing antibodies with limited cross-reactivity. In Aim 3 we propose to evaluate the epitope specificity of the broadly cross-reactive neutralizing antibodies by using phage display technology to isolate monoclonal antibodies from rabbits immunized for induction of broadly cross-reactive neutralization. The results should elucidate features of the immunogen and/or B cell response that limit the potency of, and the epitope broadly cross-reactive neutralizing antibodies. The results should facilitate the design of human HIV-1 vaccine.